CN109632649A - SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method - Google Patents
SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method Download PDFInfo
- Publication number
- CN109632649A CN109632649A CN201811523406.2A CN201811523406A CN109632649A CN 109632649 A CN109632649 A CN 109632649A CN 201811523406 A CN201811523406 A CN 201811523406A CN 109632649 A CN109632649 A CN 109632649A
- Authority
- CN
- China
- Prior art keywords
- neuron
- quantitative analysis
- fiber laser
- laser arrays
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1218—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing using optical methods; using charged particle, e.g. electron, beams or X-rays
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Neurology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
This application provides a kind of SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method, comprising: obtaining influences SF6The training sample of gas fiber laser arrays quantitative analysis factor;Based on feedforward neural network and training sample training neuron excitation function;Default neuroid desired output R exports neuron by training acquisition according to the neuroid desired outputClose to the weight and threshold value of the neuron of desired value R;According to the weight of the neuron of acquisition and threshold value and combine influence SF6SF is predicted in the input of gas fiber laser arrays quantitative analysis factor6Gas fiber laser arrays quantitative analysis results.SF provided by the present application based on artificial neural network6Gas fiber laser arrays quantitative analysis method, can effectively reduce the external factor including pressure, temperature and laser power influences, the characteristic of prominent twocomponent signal to be measured, it is ensured that the slickness of spectroscopic data can effectively improve the accuracy of analysis of spectroscopic data.
Description
Technical field
This application involves gas-insulated transformer equipment technical field more particularly to a kind of SF based on artificial neural network6
Gas fiber laser arrays quantitative analysis method.
Background technique
SF6Gas possesses excellent insulation performance, is gas componant important in GIS gas-insulated transformer equipment.Therefore,
For SF6The detection and quantitative analysis of gas just seem additional important.SF is being detected using fiber plant6During gas, inspection
Surveying result will receive the influence of many factors, such as laser intensity, pressure, temperature, spectrum peak intensity.
Currently, being mostly to carry out SF using Raman spectrum6Gas detection quantitative analysis.Specifically, it is known dense to first pass through measurement
The raman spectrum of the gas of degree establishes the relationship between spectrogram and concentration, then while measuring unknown concentration spectrogram can be according to it
Preceding relationship deduces concentration.SF is being carried out using Raman spectrum6When gas detection quantitative analysis, only considered spectrogram and
The corresponding relationship of concentration between the two, has ignored the influence of the external interferences factor such as pressure.However, SF6Gas concentration usually by
The influence of the external interferences factor such as laser intensity, pressure, temperature.Therefore, SF is carried out using Raman spectrum6Gas detection is quantitatively divided
The accuracy of analysis is unable to get the guarantee of high quality.
Summary of the invention
This application provides a kind of SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method, Neng Gouyou
Effect carries out intelligence computation to multiple parameters such as collected spectroscopic data and pressure, temperature, improves the accuracy of detection.
This application provides a kind of SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method, the side
Method includes:
Obtaining influences SF6The training sample of gas fiber laser arrays quantitative analysis factor;
Based on feedforward neural network and training sample training neuron excitation function;
The neuron excitation function isWherein, f (y) is the excitation letter of neuron
Number, z are that the operation of neuroid exports, and x is the input of neuron, and k is the weight of corresponding neuron input, and b is neuron
Threshold value;
Default neuroid desired output R makes nerve by training acquisition according to the neuroid desired output
Member outputClose to the weight and threshold value of the neuron of desired value R;
According to the weight of the neuron of acquisition and threshold value and combine influence SF6Gas fiber laser arrays quantitative analysis factor
SF is predicted in input6Gas fiber laser arrays quantitative analysis results.
Optionally, the above-mentioned SF based on artificial neural network6It is described according to institute in gas fiber laser arrays quantitative analysis method
Stating neuroid desired output exports neuron by training acquisitionClose to the weight and threshold value of the neuron of desired value R
In, the training is using adaptive adjusting step and the improved Back Propagation for adding factor of momentum.
Optionally, the above-mentioned SF based on artificial neural network6It is described adaptive in gas fiber laser arrays quantitative analysis method
Adjusting step includes:
Default initial step length, if error function E increases, by step-length Z multiplied by a constant U less than 1 along former direction weight
Newly calculate next iteration point;If error function E reduces after an iteration, then step-length Z is greater than to 1 constant h multiplied by one, is added
Big training paces;
Wherein, the error function isN is training sample sum, and m is input neuron
Number.
Optionally, the above-mentioned SF based on artificial neural network6It is described adaptive in gas fiber laser arrays quantitative analysis method
Adjusting step and the improved Back Propagation for adding factor of momentum, comprising:
Wherein, k is learning rate, n0For the number of iterations, E is
Error function, Δ k are momentum change amount.
Optionally, the above-mentioned SF based on artificial neural network6In gas fiber laser arrays quantitative analysis method, the influence SF6
The factor of gas fiber laser arrays quantitative analysis includes spectrum peak intensity, pressure, temperature and laser power.
SF provided by the present application based on artificial neural network6Gas fiber laser arrays quantitative analysis method, can be effectively right
The multiple parameters such as collected spectroscopic data and pressure, temperature carry out intelligence computations so that obtained SF6 gas concentration data compared with
The data that unused artificial neural network algorithm obtains are more accurate, effective.It is provided by the present application based on artificial neural network
SF6Gas fiber laser arrays quantitative analysis method, can effectively reduce the external factor including pressure, temperature and laser power
It influences, the characteristic of prominent twocomponent signal to be measured, it is ensured that the slickness of spectroscopic data improves the accuracy of spectral signal, can have
Effect improves the accuracy of analysis of spectroscopic data.SF provided by the present application based on artificial neural network6Gas fiber laser arrays are quantitative
Analysis method improves the accuracy of the method by Raman map detection gas concentration, considers the external interferences factor shadows such as pressure
It rings, and eliminates as much as the influence of its quantitative analysis results.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without any creative labor,
It is also possible to obtain other drawings based on these drawings.
Fig. 1 is the SF provided by the embodiments of the present application based on artificial neural network6Gas fiber laser arrays quantitative analysis method
Structure flow chart;
Fig. 2 is the structure chart of feed forward neural metanetwork provided by the embodiments of the present application.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
As shown in Fig. 1, the SF provided by the embodiments of the present application based on artificial neural network6Gas fiber laser arrays quantitatively divide
Analysis method, comprising:
S101: obtaining influences SF6The training sample of gas fiber laser arrays quantitative analysis factor.
S102: based on feedforward neural network and training sample training neuron excitation function;
The neuron excitation function isWherein, f (y) is the excitation letter of neuron
Number, z are that the operation of neuroid exports, and x is the input of neuron, and k is the weight of corresponding neuron input, and b is neuron
Threshold value.
S103: default neuroid desired output R, made according to the neuroid desired output by training
Neuron outputClose to the weight and threshold value of the neuron of desired value R.
S104: according to the weight of the neuron of acquisition and threshold value and influence SF is combined6Gas fiber laser arrays quantitative analysis
SF is predicted in the input of factor6Gas fiber laser arrays quantitative analysis results.
Test obtains the spectrum peak intensity of gas in GIS gas-insulated transformer equipment, pressure, temperature, laser power and right
The SF for the test answered6Gas concentration.
In the embodiment of the present application, the feedforward neural network of use is made of input layer, hidden layer and output layer, it can lead to
It crosses and known gas sample is learnt and grasps experience, to make differentiation to unknown gas sample.The characteristic of whole network
The threshold value being decided by the connection weight and neuron of the adjacent interlayer neuron of network.The structure of each neuron in hidden layer and output layer
As shown in Fig. 2, x in Fig. 21, x2..., xnFor the input of neuron, k1, k2..., knFor the weight accordingly inputted, b is neuron
Threshold value, f (y) be neuron excitation function, z be neuroid operation export.
In the embodiment of the present application, excitation function can choose Log-sigmoid function.
WithIndicate the operation output of entire neural network, R indicates desired output, then can make mind by training algorithm
Power and threshold value through network are determined under certain criterion, to make close to desired value R.
In the embodiment of the present application, used training algorithm is adaptive adjusting step and the improvement BP for adding factor of momentum
Algorithm.In the training process of network, the convergence of network can be accelerated according to the trend adjust automatically training pace of error function
Speed shortens the training time.
If an initial step length, if error function after an iteration(N is that training sample is total
Number, m are output neuron number,It is exported for the operation of j training sample, RjFor the desired output of j training sample) increase, then will
Step-length Z recalculates next iteration point along former direction multiplied by a constant U less than 1;If error function E subtracts after an iteration
It is small, then step-length Z is greater than to 1 constant h multiplied by one, increases training paces.That is:
Z(n0)=Z (n0- 1) U works as Δ E > 0
Z(n0)=Z (n0- 1) h works as Δ E < 0
Wherein, n0Indicate the number of iterations, Δ E=E (n0)-E(n0-1)。
Add the training algorithm of factor of momentum that can reduce network for the sensitivity of error surface local detail, effectively
Network is inhibited to sink into local minimum.Its thought is the modification direction using factor of momentum T memory last moment power: T Δ k (n0),
And this trend is taken into account in subsequent time.
Adaptive adjusting step and plus the algorithm for training network of factor of momentum can be represented by the formula:
Wherein, k is learning rate, as adaptive learning speed
Learning rate in the adjustment formula of rate;n0The n-th step for the number of iterations, i.e., in iteration;E is error function, i.e. the mistake of the n-th step
Poor quadratic sum;Δ k is the momentum change amount in additional guide vanes.
Additional guide vanes and adaptive learning learning rate method are that BP algorithm learning efficiency is low, convergence rate is slow for reply, and
The improved method for being easily trapped into local minimum state and being arranged.
Above-mentioned specific example is the training to the weight of neuroid, similar with its to the training of neuron threshold value,
This is repeated no more.
In the embodiment of the present application, preferred parameter T, U, h take 0.9,0.7,1.05 respectively.
The 30 groups of gas samples obtained by experiment are divided into two parts, take 20 groups (to teach for network training wherein appointing
Teacher), remaining 10 groups for predicting.In view of applicable principle, feedforward neural network used has a hidden layer.Input layer mind
It is equal to through first number and inputs influence factor number in need of consideration, i.e. peak intensity, pressure, temperature and laser power, hidden nodes
4 are taken as, the SF that output layer neuron number is6Gas concentration.In network, the excitation function of neuron in hidden layer and output layer
Choose Log-sigmoid function.
Based on the above rule, recognition result of the neuroid on 10 groups of forecast samples is shown in Table one:
Table one:
As shown in Table 1, calculated result and actual result are very nearly the same, it was demonstrated that the intelligence of artificial neural network used in the present invention
The scheme that algorithm surveys SF6 gas concentration is reliable, and error is within the allowable range.
SF provided by the embodiments of the present application based on artificial neural network6Gas fiber laser arrays quantitative analysis method, can
Intelligence computation effectively is carried out to multiple parameters such as collected spectroscopic data and pressure, temperature, so that obtained SF6 gas concentration
Data are more accurate, effective compared with the data that artificial neural network algorithm obtains are not used.It is provided by the embodiments of the present application to be based on people
The SF of artificial neural networks6Gas fiber laser arrays quantitative analysis method, can effectively reduce and exist including pressure, temperature and laser power
Interior external factor influences, the characteristic of prominent twocomponent signal to be measured, it is ensured that the slickness of spectroscopic data improves spectral signal
Accuracy can effectively improve the accuracy of analysis of spectroscopic data.SF provided by the embodiments of the present application based on artificial neural network6
Gas fiber laser arrays quantitative analysis method improves the accuracy of the method by Raman map detection gas concentration, considers pressure
Etc. external interferences factor influence, and eliminate as much as the influence of its quantitative analysis results.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment it
Between same and similar part may refer to each other, each embodiment focuses on the differences from other embodiments,
The relevent part can refer to the partial explaination of embodiments of method.Those skilled in the art are considering the hair of specification and practice here
After bright, other embodiments of the present invention will readily occur to.This application is intended to cover any modification of the invention, purposes or fit
Answering property changes, these variations, uses, or adaptations follow general principle of the invention and do not invent including the present invention
Common knowledge or conventional techniques in the art.The description and examples are only to be considered as illustrative, the present invention
True scope and spirit be indicated by the following claims.
It should be understood that the application is not limited to the precise structure that has been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims.
Claims (5)
1. a kind of SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method, which is characterized in that the method packet
It includes:
Obtaining influences SF6The training sample of gas fiber laser arrays quantitative analysis factor;
Based on feedforward neural network and training sample training neuron excitation function;
The neuron excitation function isWherein, f (y) is the excitation function of neuron, and z is
The operation of neuroid exports, and x is the input of neuron, and k is the weight of corresponding neuron input, and b is neuron threshold value;
Default neuroid desired output R keeps neuron defeated according to the neuroid desired output by training acquisition
OutClose to the weight and threshold value of the neuron of desired value R;
According to the weight of the neuron of acquisition and threshold value and combine influence SF6The input of gas fiber laser arrays quantitative analysis factor,
Predict SF6Gas fiber laser arrays quantitative analysis results.
2. the method according to claim 1, wherein described pass through instruction according to the neuroid desired output
Practicing to obtain exports neuronClose in the weight and threshold value of the neuron of desired value R, the training is using adaptive adjustment step
Improved Back Propagation that is long and adding factor of momentum.
3. according to the method described in claim 2, it is characterized in that, the adaptive adjusting step includes:
Default initial step length counts step-length Z multiplied by a constant U less than 1 if error function E increases again along former direction
Calculate next iteration point;If error function E reduces after an iteration, by step-length Z multiplied by a constant h greater than 1, instruction is increased
Practice paces;
Wherein, the error function isN is training sample sum, and m is input neuron number.
4. according to the method described in claim 3, it is characterized in that, the adaptive adjusting step and the improvement for adding factor of momentum
BP algorithm, comprising:
Wherein, k is learning rate, n0For the number of iterations, E is error
Function, Δ k are momentum change amount.
5. the method according to claim 1, wherein the influence SF6The factor of gas fiber laser arrays quantitative analysis
Including spectrum peak intensity, pressure, temperature and laser power.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811523406.2A CN109632649A (en) | 2018-12-13 | 2018-12-13 | SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811523406.2A CN109632649A (en) | 2018-12-13 | 2018-12-13 | SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109632649A true CN109632649A (en) | 2019-04-16 |
Family
ID=66073487
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811523406.2A Pending CN109632649A (en) | 2018-12-13 | 2018-12-13 | SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109632649A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110146455A (en) * | 2019-05-27 | 2019-08-20 | 安徽大学 | Laser spectrum gas concentration measuring method based on deep learning |
CN113758520A (en) * | 2021-08-18 | 2021-12-07 | 贵州众创仪云科技有限公司 | Internet of things laboratory environment monitoring method and system based on neural network |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103150594A (en) * | 2013-03-18 | 2013-06-12 | 山东电力集团公司济宁供电公司 | Transformer fault diagnosis method based on added momentum item BP (back propagation) neural network |
CN106324459A (en) * | 2016-09-30 | 2017-01-11 | 国家电网公司 | GIS local discharging detecting system and method based on sound-electricity integration sensor |
CN108107716A (en) * | 2017-12-19 | 2018-06-01 | 电子科技大学 | A kind of Parameter Measuring method based on improved BP neural network |
-
2018
- 2018-12-13 CN CN201811523406.2A patent/CN109632649A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103150594A (en) * | 2013-03-18 | 2013-06-12 | 山东电力集团公司济宁供电公司 | Transformer fault diagnosis method based on added momentum item BP (back propagation) neural network |
CN106324459A (en) * | 2016-09-30 | 2017-01-11 | 国家电网公司 | GIS local discharging detecting system and method based on sound-electricity integration sensor |
CN108107716A (en) * | 2017-12-19 | 2018-06-01 | 电子科技大学 | A kind of Parameter Measuring method based on improved BP neural network |
Non-Patent Citations (8)
Title |
---|
IRAWAN R. ET AL.: "Continuous monitoring of SF6 degradation in high voltage switchgear using Raman scattering", 《IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION》 * |
中华医学会: "《临床技术操作规范 激光医学分册》", 30 April 2010 * |
刘建娟等: "BP神经网络改进算法及其应用", 《INTELLIGENT INFORMATION TECHNOLOGY APPLICATION ASSOCIATION.PROCEEDINGS OF 2011 INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTATION AND INDUSTRIAL APPLICATION(ICIA 2011 V1)》 * |
宋福印等: "基于BP神经网络的水蒸气红外透过率仿真", 《光电子·激光》 * |
李志刚: "《光谱数据处理与定量分析技术》", 31 July 2017 * |
石春燕等: "人工神经网络在混合气体定量检测中的应用", 《仪器仪表与检测技术》 * |
陈远鸣等: "基于改进型BP神经网络的SF6气体传感器", 《电子测量与仪器学报》 * |
高峰等: "基于人工神经网络的气体定性分析方法", 《青岛大学学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110146455A (en) * | 2019-05-27 | 2019-08-20 | 安徽大学 | Laser spectrum gas concentration measuring method based on deep learning |
CN113758520A (en) * | 2021-08-18 | 2021-12-07 | 贵州众创仪云科技有限公司 | Internet of things laboratory environment monitoring method and system based on neural network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107247989B (en) | Real-time computer vision processing method and device | |
CN108876021B (en) | Medium-and-long-term runoff forecasting method and system | |
Firat et al. | Wind speed forecasting based on second order blind identification and autoregressive model | |
CN104504475A (en) | AR*-SVM (support vector machine) hybrid modeling based haze time series prediction method | |
CN109632649A (en) | SF based on artificial neural network6Gas fiber laser arrays quantitative analysis method | |
CN102809718A (en) | Ultra-high-frequency partial discharge signal identification method for gas insulated switchgear (GIS) | |
CN104063577B (en) | Method for forecasting characteristic gas development tendency in transformer oil based on generalized recurrent neural network | |
CN105631550A (en) | Short-term wind power combined prediction method | |
CN106529724A (en) | Wind power prediction method based on grey-combined weight | |
CN110969282A (en) | Runoff stability prediction method based on LSTM composite network | |
CN110186533A (en) | A kind of short-term tide prediction method in high-precision river mouth | |
CN114329347A (en) | Method and device for predicting metering error of electric energy meter and storage medium | |
WO2024109954A1 (en) | Mineralization coefficient prediction method and apparatus for sandstone-type uranium ore | |
CN108183754B (en) | Electromagnetic radiation prediction method for GSM base station | |
US20210012247A1 (en) | Information processing apparatus, information processing method, and program | |
CN110163332B (en) | Transformer fault diagnosis method | |
CN109697630B (en) | Sparse regression-based merchant passenger flow volume multi-factor analysis and prediction method | |
CN109582915A (en) | The non-linear observability degree adaptive filter method of improvement applied to bearingsonly tracking | |
CN106326982B (en) | A kind of more meteorological wind speed fusion methods optimized based on neural network and pattern search | |
CN115907178A (en) | Clean ecosystem CO 2 Method for predicting exchange amount | |
CN105277852A (en) | Classification and identification method of line conditions of power distribution network | |
CN115375181A (en) | Intelligent analysis method and system for laboratory environment quality | |
CN115270636A (en) | Deep learning-based marine fuel oil trace element analysis method | |
Glynn et al. | Estimation and inference for non-stationary arrival models with a linear trend | |
CN106709587B (en) | Direct radiation prediction method based on conventional weather forecast |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190416 |